DC to DC telecom PDU

Help us decide what to build next.
User avatar
LRL
Experienced Member
 
Posts: 220
Joined: Sun Nov 23, 2014 4:00 am
Location: Rock Springs, WY
Has thanked: 18 times
Been thanked: 46 times

DC to DC telecom PDU

Tue Jan 29, 2019 3:19 am

As we've started to deploy more BIG carrier type radios one of the struggles is powering them. Yes we can deploy a regular -48v teleco UPS and power these radios from it, but most reasonably priced (cough not) don't give you the options to power cycle a radio or turn it off remotely.

Our home-brew has been to use a RB750 or a Netonix to actuate relays to do the basics of on or off. But it could be so much more helpful to have power draw info with a nice configurable power UI like the DC switches.

Honestly a similar concept to UBNTs edge power, but with at least 4 DC output terminals. The way we build our sites a variable DC input would be preferred but I could see some pretty good potential for an AC option. You could actually get really fancy and integrate this all into a reasonably priced telco UPS (with cust supplied batteries). Some of the radios we've been using are power hungry though, 130-140 watts. When you start using multiple RTUs on the same dish with layer1 aggregation powering these types of links messy.

I've had a lot of discussions with other operators and it seems to be the trend that more and more are starting to use DC switches even at AC sites and then use AC to DC converters to power and charge their own battery bank. This really makes a ton of sense, you get voltage monitoring of the bank and should the power be off long enough you can pre-set the switch(es) to make intelligent choices to power down radios to limp the site along. Or in the case of all CPEs being powerless, turn APs off to keep backhauls up longer. Add multiple AC to DC converters and suddenly you have yourself a pretty solid DC UPS and for a fraction of what's commercially sold.
-LRL

"My reading of history convinces me that most bad government results from too much government." - Thomas Jefferson

User avatar
Dave
Employee
Employee
 
Posts: 451
Joined: Tue Apr 08, 2014 6:28 pm
Has thanked: 1 time
Been thanked: 82 times

Re: DC to DC telecom PDU

Tue Jan 29, 2019 9:20 pm

LRL

What voltage/current range would each of the DC output terminals need to be?

Dave

User avatar
LRL
Experienced Member
 
Posts: 220
Joined: Sun Nov 23, 2014 4:00 am
Location: Rock Springs, WY
Has thanked: 18 times
Been thanked: 46 times

Re: DC to DC telecom PDU

Wed Jan 30, 2019 2:39 am

Dave,

I had the silent internal argument about voltage, but honestly I can't see a need for anything but 48~ volt. For resistance (voltage drop) probably a little higher (49-50) as I can't remember a radio that does have a range of at least 45-50. We've got some radios that are right under 150 watts, so I suspect 3amp would be enough, but 4 gives some wiggle room for long runs and such ;)
-LRL

"My reading of history convinces me that most bad government results from too much government." - Thomas Jefferson

User avatar
highlands
Member
 
Posts: 47
Joined: Fri Oct 09, 2015 1:01 pm
Has thanked: 4 times
Been thanked: 2 times

Re: DC to DC telecom PDU

Wed May 15, 2019 9:54 am

Good day,

Here is a suggestion. ports that are PoE only. i.e. They do not connect, data wise, to the switch. There would be a powered PoE port and a data port with all 4 power types. The logic of the Netonix would only be able to apply power and monitor power. The Netonix would not be data aware on those 2 ports as there would no connection to the switch. This allows for a connection to a backhaul radio that needs to connect to a router first and then to a switch for distribution. 2 ports would give a in and out or a upstream and a down stream. Sure maybe 4 of these on a 24 port switch and 2 on a 12 port.

As to the separate DC terminals, there really is no need for the power to come from the power supply inside the Netonix. If a separate power could be externally applied to a small power buss of the Netonix then that same 48VDC (or whatever voltage) can be separate and all the DC terminals could use a simple DPST 10A relay could be used. I would think on these ports keep them isolated so that -48VDC or +48VDC or +12VDC, +24VDC could be use and its transparent to the Netonix as it is only a power pass through with monitoring and a relay. It keeps it versatile without creating multiple models. The only slightly odd part is your current shunts have to pass current either way. This would give a + or - current direction, however I don't think that would be a hard one to deal with. I would think the same count as above, maybe 4 of these on a 24 port switch and 2 on a 12 port.

Your thoughts please
JCS

Return to What should we build next

Who is online

Users browsing this forum: No registered users and 1 guest